www.cetljournal.co.uk
Written Discourse Analysis and its Application in ELT
Jianli Yao
Henan Economy & Trade Vocational College, China. E-mail: [email protected]
Abstract
This paper, by reviewing relative literature and scholarly discussion, looks at the
practicality and possibility of applying discourse analysis to English language teaching to
learners from non-English speaking countries. It then explores aspects and dimension a
written discourse need to cover in order to be helpful with task-based language teaching.
Taking an English textbook article as an example, it further illustrated how discourse
analysis may be applied to such a task-based course design to help learners with
understanding and using of the language in a natural conversational perspective. By
drawing examples of the author’s own teaching experience, the author concludes that
discourse analysis, particularly the analysis of text pattern should play a more important
part in reading and writing lessons alike in ELT.
Key words: discourse analysis, written discourse, English language learning, TEFL
Introduction
Discourse analysis of language, spoken and written alike, proved to be very helpful
and inspiring in linguistics studies and English language teaching (ELT). This paper, with a
specific focus on analysis of written texts, deals with some basic aspects of discourse
analysis and its possible application in ELT. By making a simple literature review at the
beginning, the first part makes a brief introduction to the origin and development of
discourse analysis as a discipline. And then, the first part mainly covers some elementary
conceptions and dimensions to be taken into account when the written discourse is analyzed.
The second part in turn discusses a text entitled Earthquakes Leave Loose Material All Shook
Up, taken from New Scientist, Volume 126. This part makes an illustration in detail of
different aspects of written discourse analysis listed in previous part, particularly an analysis
42
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
www.cetljournal.co.uk
of its text patterns and its selection of cohesive devices. Lastly, by referring to the analysis
done in part two, some suggestions are raised about how analysis of written discourse could
help in teaching written English, with examples of my own teaching experience.
What is Meant by Discourse Analysis and What to Analyze in a Written
Discourse?
As McCarthy, M. (1991) summarized, when most linguists’ major concerns were still
with analyzing the structure of sentences, Zellig Harris published his paper entitled
Discourse Analysis in 1952, in which he showed interests in the linguistic elements
distribution in extended texts. Although what he studied was different from the discourse
analysis studied today, more and more scholars, either of linguistics or of other disciplines,
began to involve themselves in relevant studies. It was from all those studies in 1960s and
1970s, that that discourse analysis, which “is concerned with the study of the relationship
between language and the contexts in which it is used”(McCarthy, M. 1991P5), developed
into “a wide-ranging and heterogeneous discipline, which finds its unity in the description of
language above the sentence and an interest in the contexts and cultural influences which
affect language in use” (McCarthy, M.1991 P7). Stubbs (1983) defined discourse analysis as
a field of research, which is concerned with 1) the use of language over the level of a
sentence/utterance, 2) the interrelationships between language and society and 3) the
interactive properties of daily communication.
In addition to M.A.K. Halliday’s functional approach to language, Sinclair and
Coulthard at the University of Birmingham were as important and influential to the
development of Discourse Analysis in Britain (McCarthy, M.1991). Michael Hoey also
contributed his own understanding of discourse. He roughly summarized discourse as any
stretch of spoken or written language, longer than one sentence, which is self-contained in a
reasonable way. Therefore, Hoey argued that “discourse analysis is the area of linguistics that
concerns itself with the study of these multi-utterance acts of communication.”(Hoey,
M.1991)
As Discourse Analysis covers both spoken interactions and all kinds of written and
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
43
www.cetljournal.co.uk
printed texts: “newspaper articles, letters, stories, recipes, instructions, notices, comics
billboards, leaflets pushed through the door, and so on.”(McCarthy, M.1991 P12), it is
impossible to talk about it in detail in a short essay like this. This essay is merely concerned
with the analysis of written discourse.
While the spoken discourse analysis focuses on the discussion of exchange structures
and analysis of conversations, the discourse analysis of written texts usually takes into
accounts coherence, cohesions and text patterns. According to Halliday and Hasan (1976), a
text, “not just a string of sentences”, can be either spoken or written and of any length. It is
not simply a large grammatical unit, something of the same kind as a sentence, but a
semantic unit whose “texture” is dictated by its interpretation within a particular context, or
environment. Halliday and Hasan (1976) have done much research into what makes a text a
text, i.e. how we can differentiate a cohesive grammatical unit from a random collection of
sentences. Five cohesive devices have been sorted out, namely, “reference, substitution,
ellipses, conjunctions and lexical ties” (Hatch, 1992 p223).
Reference is usually established by using pronouns, demonstratives and comparatives
as cohesive ties. If subdivided, reference could be exphoric when referring to something
outside the text, or it can be endophoric, a reference within a text, referring to the person(s)
or item(s) talked about within a previous (anaphoric) and/or succeeding (cataphoric) context.
This can be more easily recognized through a grid as follows:
Figure 1: Establishment of a Reference within a Text
|-Exphoric
Reference-
|Anaphoric
|-Endophric
|Cataphori
c
In contrast
to reference, substitution is the replacements of an item mentioned
previously. It can be used to substitute nominal, verbal or clausal items. For example, when
an item is mentioned for the second time, it is more likely to be replaced by one(s) or it (them)
to avoid unnecessary repetition.
Ellipses, seemingly the same as substitution, are also used to establish ties to
44
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
www.cetljournal.co.uk
nominals, verbals and clauses for the sake of concision. What distinguishes ellipsis from
substitution is that ellipsis is a “zero” cohesive devices because it is not actually said or
written down.
Another cohesive device, conjunction, as its name suggests, is employed to link
clauses, such as besides, yet, therefore, then, etc, by showing additive, adversative, causal,
temporal or other different kinds of conjunctive relations.
The last category of cohesive device of Halliday and Hasan’s system is the device of
lexical ties. To achieve lexical cohesion, we can use repetition, synonym, near synonym,
superordinate, general words, antonym ordered series, metonymy, members of the same
lexical set or any words from the same semantic field (Halliday and Hasan, 1976, chapter 6).
Besides grammatical and lexical cohesion, clausal relations and text patterns should also be
dealt with in written discourse analysis. Clausal relations, including logical sequence,
matching or multiple clause relations, are defined as shared cognitive processes “whereby we
interpret the meaning of a clause or group of clauses in the light of their adjoining clause or
group of clauses” (Winter 1994:p49). Text patterns are something to show the writer’s logic
of organizing ideas and his/her actual way of presenting them in a written text since “every
writer is face with the problem of how or organize and present his/her non-linear message in
comprehensible linear form” (Coulthard, M. 1994 p7). As Hoey (1983 and 2001)
summarized, text patterns fall into categories including: a). problem-solution pattern, through
which a solution is provided to a problem raised by the writer, usually at the beginning of the
text; signaled by words as issue, problem, situation, and assessment, approaches, solution; b).
goal-achievement pattern, showing the relationship between what people set as aims and the
way to realize them, with signaling words like want to, would like to, aim, objective and
means, method, way; c). opportunity-taking pattern, providing chances in certain situations
and how to make use of them, using signals including opportunity, offer, unique, special,
unusual, outstanding and meet, come upon, find, read, hear, or see; d). gaping in knowledgefilling, providing concrete and exact answer to what may be beyond people’s common sense,
signaled by words such as question, puzzle, mystery and explanation, hypothesis, theory,
suggest, solve; e). claim-counterclaim (response) pattern, whose signals are often words of
claim, and words of denial or affirmation, usually signaled by claim, suggest, propose and so
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
45
www.cetljournal.co.uk
on; or f). Interlocking pattern, which may combine goals-methods and problem-response in
one. It is through all these “culturally popular” text patterns that the authors present the
readers cohesive, coherent, interesting, inspiring, and/or thought-provoking written
discourses.
Analysis and Discussion of Earthquake Leave Loose Material All Shook Up
After discussing the dimensions covered by written discourse analysis, I will analyze
a text entitled Earthquake Leave Loose Material All Shook Up, written by Jeff Hecht and
published in the magazine New Scientist, Volume 126, in April 1990. Through analyzing its
selection of and text patterns, I hope to make clearer my discussion about written discourse
analysis conducted in part one.
First of all, I would like to deal with the cohesive devices used in this article. As
everyone can imagine, reference is the most frequently used cohesive device in English,
particularly the one established through pronouns and demonstrative. The sample passage
provided us many an example of references realized in this way. For example, the pronoun
they in sentences3 in paragraph I, they in sentence 2,3 in paragraph II and they in sentence2,
3 in paragraph III all refer back to mounds. To be more exact, we can see them as examples
of anaphoric reference as they refer to mounds in previous senctence(s). Similarly, they in
sentence 2 of paragraph VII refers to particles in sentence 1 while the very last word them
used to refer to mounds again. Other reference types can also be found as word similar in
sentence1 of paragraph V is an example of comparative reference, pointing at the way that
ash to clump together to form miniature mounds. And such in sentence 4 of paragraph VI and
here in sentence 2 of paragraph VII are demonstrative reference, referring respectively to
standing waves (sentence 3, paragraph VI) and null zones (sentence1, paragraph 7).
Interestingly, there is one more such example, as this in the second sentence below the
illustration referring to what is mentioned in previous sentence.
Besides, the author also used ellipses in the article, as can be seen in Charles Higgins
of the university of California at Davis says: “I don’t think …it explains …that many
published hypotheses don’t” sentence 2 of paragraph VIII--- the verb explains is omitted after
don’t. In addition, in sentence 2 of paragraph IX, others, is an example of substitution, taking
46
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
www.cetljournal.co.uk
the place of mounds. All the but in sentence 4 of paragraph III, in sentence 3 of paragraph V
and in the very last sentence of the passage are examples of conjunction. Other examples are,
while in sentence 2 of paragraph IV, when in sentence 5 in paragraph V, and however in
sentence 2 of paragraph IX.
Furthermore, many examples of lexical cohesions can be found throughout the
passage. For instance, mounds are simply repeated so many times all through the article
while hillock used as its synonymy in paragraph I. In paragraph II and III, words with similar
meanings like forming, existence, built, and result from help realize lexical cohesion in
talking about the origin of mounds. And From paragraph VI through IX, the author employed
general words like places zones and areas to achieve cohesion in lexis when talking about the
sites where mounds can be found.
In addition to analyzing cohesive devices, clausal relations and text patterns of a
higher textual level are to be analyzed in turn. Good writers need take into account the
readers when they write (Koester, 2002 personal communication). Such a principle makes it
very important for us to pay some attention to aspect. As mentioned in part one of this essay,
clausal relations include logical sequence, matching or multiple clause relations. When it
comes to this aspect, lexical cohesion device of conjunctions also play a rather important role
and should be mentioned once again, because logical sequence relations are always signaled
by conjunctions, though they do not have to be (Koester, 2002 personal communication). It is
a shame that in the case of the analyzed text, we can only observe conjunctions like but and
however are used to show contrast clausal relation, one type of matching relations between
clauses.
Fortunately, this passage, as a piece of scientific writing, is a good example of claimcounterclaim (response) pattern of text, by the hint of signaling words like believe in
sentence 2 and suggests in sentence 3 of paragraph I, proposed in sentence 1, claim in
sentence 2 of paragraph II and many others. According to Hoey (2001), claim-counterclaim
(response) pattern, also known as hypothetical-real pattern (Winter, 1994), which can be
illustrated as shown in figure 2:
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
47
www.cetljournal.co.uk
Figure 2: Stages in Claim-counterclaim Patterns, Based on Hoey (2001) p180
The general pattern structure of the cited passage, therefore, can be analyzed as:
Situation: Mysterious Origin of Mima Mounds
Claim: the vibration of earthquakes might cause Mima Mounds, made by Andrew Berg
Reasons for claim:
 his incidental experience of seeing the effect of vibration, (paragraph IV)
 no any similar previous suggestions by other geologists (paragraph V)
 hypothesis about the effect of seismic waves caused by earthquake (paragraph VI)
Affirm and Reason for it:

many mounds are in earthquake zones (sentence 1, paragraph IX)

explained aspects failed to be mentioned by other hypothesis (sentence 2 in
paragraph VIII)
Denial / Negative Evaluation:
 no quantitative model of this phenomenon (sentence 2, paragraph VII)
 not predictable amount of energy to be needed (sentence 3, paragraph VII)
 other geologists’ support of gopher theory
48
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
www.cetljournal.co.uk
Counter Claim(s):

gopher theory (paragraph III)

product of selective erosion of soil (paragraph III)

effect of glacial deposition (paragraph III)

accumulations of soil around roots of plants (paragraph III)
As there is not yet an agreement reached among all the geologists, this article does
not provide any correction to the claim and therefore the reason for correction is also missing.
Therefore, if the relevant information of this passage is added, the pattern shown in
Figure 2 can be modified as follows. The words in double-line-edged grids attached to each
item are listed as the explanation shown in Figure 3:
Figure 3: Modification of Claim-counterclaim Patterns
As different readers may interpret the text patterns from various angles, Berg’s claim
could be analyzed as counterclaim against other geologists’ suggestions if theirs are regarded
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
49
www.cetljournal.co.uk
as claims. It can also be argued that, if the origin of mounds is analyzed as a question raised
by the author, all the claims of Berg or other geologists can be analyzed as possible solutions.
Then the text pattern of the sample passage will fall into to another category, the problemsolution pattern, “arguably the most common pattern of all” (Hoey, 2001 p123). But just as
Hoey (2001) argued, sometimes it is not necessary to distinguish whether a certain passage
has this or that pattern since “clearly all the patterns we have been considering have much in
common”(p166). They can all be summarized as SPRE (situation, problem, response and
evaluation) pattern, and the part of problem, in many circumstances can also be the part of
goal, opportunity, desire arousal, gap in knowledge, which triggers different responses, which
then in turn lead to a negative evaluation to recycle the pattern and finally a positive
evaluation to end the pattern.
Application of Written Discourse Analysis in ELT
Needless to say, analysis of written discourse is not just for fun of a very challenging
intellectual game. McCarthy (1991) concluded that, though “not everything described by
discourse analysis is relevant to or may have immediate applications in language teaching”
(McCarthy, 1991. p147), language teachers are more likely to be able to “create authentic
materials and activities for the classroom” (McCarthy, 1991. p147) if they are more familiar
with how different texts are organized and how to realize the process of creating written texts.
Complete the gaps in the appropriate pronoun: I, you, he, she, it, we, they
The parents of a seven-year-old Australian boy woke to find a giant python crushing
and trying to swallow ____.
The incident occurred in Cairns, Queensland and the boy’s mother, Mrs. Kathy Dryden
said, “_____ was like a horror movie. It was a hot night and Bartholomew was lying
under a mosquito net. _____ suddenly started screaming.
“____ rushed to the bedroom to find a huge snake trying to strangle ____. ____ was
coiled around his arms and neck and was going down ___ body.
Mrs. Dryden and her husband Peter, tried to stab the creature with knives but the python bit the boy several times before escaping.
50
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
www.cetljournal.co.uk
According to what is discussed in previous parts, it can be concluded that
understanding of cohesive devices will be quite helpful for language learners. In reading
practices, especially when doing exercises designed to increase reading speed, it is very
important for the readers to see, without referring back to the read paragraphs, who are the
referees whenever they encounter pronouns and/or demonstratives. Language teachers should
provide constant guidance to the learners and always remind them of paying enough attention
to this aspect. Besides, task-based activities may be designed to help students with their
vocabulary study, through more understanding of lexical cohesion. The students do not have
to stop to consult the dictionary if they can take a good guess at the meaning of some new
words through their knowledge of lexical cohesions, especially those, particularly when
synonyms, antonyms and/or words of the same semantic field are used. For example, in an
exercise as above.
As Hatch (1992) cited, Johns (1984) noted that lexical cohesions are not as frequently
used by Chinese teachers as by native speakers, since words and phrases are taught not as
part of a semantically related chain but as isolated items. Things have changed a lot in ELT in
China as traditional audio-lingual is giving its place to task-based or teaching. Vocabulary,
due to its huge amount, tends to be the biggest problem for most of my students.
Unfortunately, they should be told to remember new words in such a way so as to achieve a
higher efficiency. Therefore, besides trying to find out the pronoun reference, students can
also be guided to find out the meaning of the word python, with the help of its superordinate
snake, after which they can be guided to find out words relevant to snake, including crushing,
swallow, coiled around and bit.
In addition to employing cohesive devices in reading classes and/or vocabulary study,
analysis of text pattern, in my personal opinion, is of an even greater importance. Hoey (2001)
provided some implications of written discourse analysis for language learners, particularly
those, like my former students, who are learning a language with a very different “rhetorical”
expectations. Such learners should be instructed about ways “various patterns operate”. My
students in China are mainly college students with an intermediate or higher level of English.
And according to my personal understanding and experience, the analysis of text pattern is of
particular importance for them. The students should have some knowledge of those culturally
popular text patterns so that they will have the idea of what to expect in a certain passage and
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
51
www.cetljournal.co.uk
they will also know what they are expected to include when making a certain piece of writing,
since reading skills and writing techniques are parts of the same “organic body”. However, it
is a shame that, in the practice of teaching English to college students in China, reading and
writing are usually given, and therefore taken by the students, as two separated subjects.
According to my own experience, such an approach reduces the mutual effect of reading and
writing classes, whose mutual effects should otherwise be enhanced. What I usually do is to
try to combine the instruction of reading skills and writing techniques into a whole coherent
process. With the help of the analysis of written discourse, particularly that of text pattern,
this goal is easier to be achieved.
For example, through a passage like Earthquake Leave Loose Material All Shook Up,
a task-based lesson may be designed as follows to combine reading and writing practice into
one by doing the discourse analysis.
To begin with, the teacher can have a discussion with the students about some
mysteries in nature. This discussion is both a brainstorming and warming-up exercise, during
which the teacher introduces the topic, the origin of the mounds. Then the teacher should
give his/her students several minutes to scan the whole passage, which is then given to them
as handouts for them to find some more words concerning this topic, or put it another way,
words of the same semantic field.
After dealing with the new words in this passage, the teacher should then ask the
students read the passage carefully to detect all the different claims made by different
geologists about the origins of the mounds. The teacher should guide the students to analyze
the text and give them the whole idea of claim-counterclaim (response) pattern by drawing a
diagram of this pattern on the whiteboard. Then the students will be asked to find out the
content of each correspondent part of the claim-counterclaim (response) pattern, as what is
done in part two of this essay, while the teacher deals with the passage through the grammartranslation or whichever approaches he/she usually employs.
When the student all have a fairly clear idea of the way a claim-counterclaim
(response) operates, they can be given a further introduction to some other text patterns and
then the generalization of SPRE pattern. That is the end of the reading and analyzing part of
the lesson.
52
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
www.cetljournal.co.uk
The last part of this lesson should be the writing exercise, in which students should
try to write an outline for a composition about a “mystery” in nature by offering some
possible explanation to it. And they should be reminded to give the situation, their own claim
and some other counterclaims. They should also provide reasons for the claims they make as
well as reasons for the counterclaims from other sources. Lastly, they should make evaluation
of the claim and reasons they present. If it is still a bit hard for the students to finish this in
class, it can be given as assignment for them to do after school. But the learners do have to be
reminded again of taking hints from the text-pattern of claim-counterclaim (response).
This is of course just a simple case of applying written discourse analysis in teaching
written English. And it should be in a greater detail when actually applied in ELT classes. But
it is definite that analysis of written discourse, especially that of text pattern, could help both
the teacher and the students out in dealing with what to expect in a reading passage and what
should be included when writing a certain passage.
Summary
From what McCarthy (1991) and Hoey (1991) have defined and summarized, it can
be concluded that discourse analysis can be very helpful in language teaching and learning as
it focuses more on naturally occurring language above the sentence level. It is certain that
language acquisition is the ultimate goal of language lessons. And the teachers expect the
student to be capable of using and understanding the language rather than recite some rules.
Therefore, discourse analysis can be especially valuable and helpful when applied to
language lessons. As Hoey (1991) mentioned, discourse analysis is relevant to language
teaching, because “if the teacher knows what a natural conversation involves, he or she will
be in a better position to assess whether their learners are succeeding in developing the
conversational skills that they need.”(Hoey, 1991) This thoroughly explains how Discourse
Analysis can help in second language teaching and learning.
Discourse analysis, particularly the analysis of text pattern should play a more
important part in reading and writing lessons alike in ELT. And it has a more significant role
to play in combing these two categories into one complete unit. From
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES
53
www.cetljournal.co.uk
REFERENCE:
Coulthard, M. 1985 An Introduction to Discourse Analysis (2nd ed.). London: Longman
Coulthard, M. 1994 On analyzing and evaluating written text. In Coulthard, M. (Eds.)
Advances in Written Text Analysis: 1-25. London: Routledge
Harmer, J. 1991 The Practice of English Language Teaching. Harlow: Longman
Hatch, E. 1992 Discourse and Language Education. Cambridge: Cambridge University Press.
Hoey, M.P. 1991 Some properties of spoken discourse. In R. Browers and C. Brumfit (Eds.)
Applied Linguistics and English Language Teaching: 65-84. Basingstoke:
Macmillan/MEP
Hoey, M.P. (1994). Signaling in discourse: a functional analysis of a common discourse
pattern in written and spoken English. In Coulthard, M. (Eds.) Advances in
Written Text Analysis: 26-45. London: Routledge
Hoey, M.P. (2001). Textual Interaction. London: Routledge
John, A. 1984 Textual cohesion and Chinese speaker of English. Language Learning and
Communication, 3, 1-92
McCarthy, M. (1991). Discourse Analysis for Language Teachers Cambridge: Cambridge
University Press
Nunan, D. (1999). Second Language Teaching & Learning Boston: Heinle & Heinle
Publishers
Polanyi, L. (1981). Discourse analysis and the role of cultural presuppositions in storytelling.
In A. James/ P. Westney (Eds.) New Linguistic Impulses in Foreign Language
Teaching (P161-183) Tübingen: Narr
Stubbs, M. 1983, Discourse Analysis, Oxford: Basil Blackwell.
54
CONTEMPORARY ENGLISH TEACHING AND LEARNING IN NON-ENGLISH-SPEAKING COUNTRIES